|Publication number||US6683552 B2|
|Application number||US 10/024,100|
|Publication date||Jan 27, 2004|
|Filing date||Dec 17, 2001|
|Priority date||Jul 17, 2001|
|Also published as||EP1415400A2, US20030016150, WO2003009478A2, WO2003009478A3|
|Publication number||024100, 10024100, US 6683552 B2, US 6683552B2, US-B2-6683552, US6683552 B2, US6683552B2|
|Inventors||John M. Noll, Brian P. Bunch, Sundara Murthy|
|Original Assignee||Honeywell International, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (11), Non-Patent Citations (1), Referenced by (10), Classifications (6), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims the benefit of U.S. Provisional Application Serial No. 60/306,195, filed in the names of John M. Noll, Brian P. Bunch, and Sundara Murthy on Jul. 17, 2001, the complete disclosure of which is incorporated herein by reference.
The present invention relates to analog-to-digital signal conversion, and in particular to dual analog-to-digital converter circuits.
Analog signals generated by various devices such as sensors are very often desired to be converted into corresponding digital signals because of the convenience and accuracy of digital signal processing. An analog-to-digital (A/D) converter is a well-known electronic circuit that receives an analog input signal and produces a digital output signal. Thus, the digital output signal is produced based on samples of the analog input signal taken over time. A/D converters are used to provide sequences of digital code representations of numbers in a chosen number base such that the numbers correspond to the amplitude value of sample portions of the analog signal input waveform with respect to a predetermined reference value.
In some circuits, it is desirable to maintain the peak input voltage of the analog signal at or near the maximum value for the range of input signals accepted by the A/D converter. Thus, the sampled analog input is typically normalized so that as much of the range of the converter is used as possible without overflowing. This normalization insures the quantization step-size is sufficiently small that sampling noise is minimized. The input signal normalization process, which is provided by a well-known Automatic Gain Control (AGC) circuit, requires both a variable gain mechanism to adjust the signal and knowledge of the gain value in downstream processing to compensate for the initial signal adjustment.
The AGC circuit typically monitors the analog input signal to the A/D converter and generates a feedback signal based on the monitored input signal. The feedback signal is provided to an amplifier to control the level of the analog input signal to the A/D converter. This feedback signal attempts to keep the peak voltage level of the analog input signal at or near the full-scale value of the input for the A/D converter.
Conventionally, the feedback signal is generated using analog circuitry. For example, such feedback loops typically include one or more of the following analog control blocks: a log amplifier, a summing amplifier, an integrator, and a differentiator. However, such feedback AGC loops are also known that utilize digital circuitry.
The present invention provides improved analog-to-digital (A/D) conversion by providing an apparatus dual channel analog-to-digital converter circuit and a method of performing dual channel analog-to-digital conversion, in contrast to the prior art devices and methods. The apparatus and method of the present invention extend the dynamic range of an A/D converter circuit and eliminates the need for an Automatic Gain Control circuit and its associated complexity.
According to one aspect of the invention, the apparatus and method of the invention are embodied as a novel dual A/D converter circuit that includes two separate channels coupled to receive a single analog input signal, each of the two channels having a means for converting an analog input signal to a digital signal at respective sample times that are adjustable to be substantially simultaneous; a means for adjusting the gain of the two channels to differ by a predetermined factor; a means for detecting a channel overflow condition in one of the two channels having a higher gain; and a means for merging the two channels into a continuous output stream.
According to another aspect of the invention, the dual A/D converter circuit of the invention includes a means for eliminating signal offset errors in the two channels. The means for eliminating signal offset errors is, for example, a digital subtraction circuit.
According to another aspect of the invention, the means for merging the two channels is operable as a function of a result generated by the channel overflow condition detecting means. Furthermore, the means for merging the two channels is operable to output a result of the channel having lower gain when a channel overflow condition is detected on the channel having higher gain.
According to another aspect of the invention, the means for adjusting the gain of the two channels further comprises an amplifier circuit in each of the two channels.
According to another aspect of the invention, the dual A/D converter circuit of the invention includes a means for matching a result of the channel having higher gain to a result of the channel having lower gain.
According to another aspect of the invention, the dual A/D converter circuit of the invention also includes a means for adjusting placement of a rising edge in a clock circuit portion of one channel in time with respect to a clock circuit portion of another channel so that the sample times of the two channels are substantially simultaneous.
According to another aspect of the invention, the channel overflow condition detecting means further comprises a digital comparator circuit.
According to still another aspect of the invention, the means for merging the two channels further comprises a means for adjusting a data bit position of a result of the channel having lower gain to match a result of the channel having higher gain.
According to yet other aspects of the invention, methods for converting an analog signal to a digital signal are provided. The methods of the invention include, for example, splitting an input analog signal into large and small signal channels; scaling the input signal on the large and small signal channels such that the small signal channel has a higher resolution than the large signal channel; sampling the large and small signal channels using separate analog-to-digital converters; and outputting a result of one of the large and small signal channels as a function of determining whether the small signal channel is valid.
The methods of the invention also include merging a result of the large signal channel with a result of the small signal channel into a merged result; and outputting the merged results.
The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same becomes better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawing, wherein:
the FIGURE is an illustration of the apparatus and method of the invention embodied as a novel dual analog-to-digital converter circuit.
The present invention is an apparatus and method for extending the dynamic range of an analog-to-digital (A/D) converter circuit that eliminates the need for an Automatic Gain Control (AGC) circuit and its associated complexity. Rather, the apparatus and method of the invention utilize two A/D converters that simultaneously sample on different channels two differently scaled versions of the same input signal. The scaling is such that one channel has a resolution that is four times higher the other channel. The higher resolution channel is monitored for data overflow, and the channel that has the highest resolution data without overflowing is selected as the output. The selected outputs are merged to produce a seamless stream of output data. The resultant output is a stream of data in which the quantization step-size is larger for large signals and four times smaller for small signals. The level of dynamic range thus provided by the present invention eliminates the implementation of traditional AGC to control the level of the analog input signal to keep the peak voltage level of the analog input signal at or near the full-scale value of the input for the A/D converter.
The FIGURE illustrates the apparatus and method of the invention embodied as a novel dual A/D converter circuit 10. As illustrated in the exemplary embodiment of the FIGURE, when sampled with a dual A/D converter circuit 10 of the invention, the input signal is split into two channels, with separate buffers dedicated for each respective channel. Respective buffer amplifiers 12 and 14 amplify the input signal on respective channels with a gain of one and a gain of four, respectively. Each channel is sampled by one of two substantially identical A/D converters 16 and 18. The two channels are sampled simultaneously. If the gain of four channel is valid, then its result is passed through unmodified as the output of the dual A/D converter circuit 10. If the gain of four channel has overflowed, then its result is discarded; rather, the result of the gain of one channel is left shifted two bit positions and used as the output. The left shift effectively adds two bits of precision to the A/D conversion of small signals, while maintaining range capacity for large signals.
Thus, according to the exemplary embodiment illustrated in the FIGURE, the dual A/D converter circuit 10 of the invention includes: a means for eliminating signal offset errors in both of two separate channels; a means for adjusting the gain of the two channels to differ by a substantially exact factor, such as a factor of four; a means for converting an analog signal input to each of the two separate channels to a digital signal at respective sample times that are adjustable to be substantially simultaneous; a means for detecting a channel overflow condition at least in the channel having the higher gain, i e., the smaller signal channel; and a means for merging the two channels on an “on the fly” basis.
Analog Signal Conditioning
The analog input signal is directed to two signal-clamping amplifier channels, a second of the two amplifier channels having a gain that is higher than the gain of the first channel by a predetermined factor. For example, the first amplifier channel has a gain of two, while the second has a gain of eight. Thus, as compared to each other, the first and second circuits differ in gain by a factor of approximately four. Clamping voltage thresholds are set at levels such that the resulting output slightly exceeds the valid range of a respective first and second channel's A/D converter 16, 18. The buffer amplifiers 12, 14 implement the voltage clamping function by comparing the input signal level and clamping the voltage if the threshold is exceeded. Alternatively, clamping diodes (not shown) can be used to clamp the voltage thresholds to achieve the desired resulting output. Each of the first and second channels is then buffered with a unity-gain single-ended to differential-output amplifier (not shown) before being directed to the respective A/D converter 16, 18.
The first and second channels are sampled by respective multi-bit high-speed A/D converters 16, 18. For example, the A/D converters 16, 18 are 14-bit high-speed A/D converters. Sample timing for the A/D converters is provided by sample clocks 24 and 26 are derived from a 64 MHz sine wave using respective differential receivers 28 and 30. One of the differential receivers 28, 30 uses a variable threshold voltage to adjust placement of the rising edge trigger in time with respect to the other receiver 30, 28 to assure that both A/D converters 16, 18 sample substantially simultaneously. The adjustable phase offset is selected using, for example, a voltage-output digital-to-analog converter (DAC) 32. During a calibration cycle, this voltage is adjusted using known means until substantially simultaneous sampling between channels is achieved.
Each A/D converter channel has an inherent DC offset error. Digital post-processing is used to modify raw samples output by the respective A/D converter 16, 18 and eliminate this signal offset error. A digital subtraction circuit 34 and 36 is included for each respective channel so that a fractional offset correction value is subtracted from the raw sample value.
Since the first and second channels differ in gain by a factor of four, a similar modification is made to the offset-corrected raw values so that the ratio between them is also a factor of four. A digital multiplication circuit 38 is included for the small-signal channel so that it is fractionally scaled to match the large-signal channel by a factor of four. The offset correction and gain scaling constants are computed during a periodic calibration cycle.
To select and format the desired output data, a digital multiplexing circuit 40 is included to merge corrected data values from both the first and second channels and present them as a single output stream. A digital comparator circuit 42 determines whether the small-signal channel has overflowed. This overflow signal from the comparator circuit 42 is used to control the multiplexer 40 and select the valid and most precise output values. To merge channels and account for the factor-of-four difference in data values, a left shift function 44 is operated that left shifts the large-signal channel data entering the multiplexer 40 by two bit positions. This left shift of the large-signal channel data effectively matches values from either channel so that the output appears as a continuous output stream. The left shift is implemented by a bus connection in at the multiplexer 40. The large signal data is connected to the higher-order bits of the multiplexer 40, leaving the two lower-order bits unused, i.e., the two lower-order bits are set to zero. The left shift function 44 is thus part of the data path of the large signal data, which eliminates any need for a shift register. However, in an alternative embodiment, the left shift function 44 may be implemented using a register that left shifts the large-signal channel data two bit positions before entering the multiplexer 40.
An optional digital comparator circuit 46 determines whether the large-signal channel has overflowed whereupon the overflow signal can be output. The large-signal channel overflow signal may be useful in down-stream processing for some applications.
While the preferred embodiment of the invention may be advantageously implemented in CMOS, with the exception of resistors and capacitors, those skilled in the art having the benefit of the disclosures of the present invention will appreciate that other semiconductor technologies may be used to implement the invention. For example, the invention may be integrated using application specific integrated circuit (ASIC) technology, gate arrays or other available integrated circuit techniques.
Furthermore, while the preferred embodiment of the invention has been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4525741 *||Nov 3, 1982||Jun 25, 1985||Ncr Corporation||Self-adjusting video camera|
|US4999628 *||Jun 15, 1990||Mar 12, 1991||Yamaha Corporation||Analog-to-digital converting unit with broad dynamic range|
|US5223932 *||Jan 10, 1991||Jun 29, 1993||Wayne State University||Dynamic offset to increase the range of digitization of video images|
|US5482036 *||May 26, 1994||Jan 9, 1996||Masimo Corporation||Signal processing apparatus and method|
|US5801657 *||Feb 5, 1997||Sep 1, 1998||Stanford University||Serial analog-to-digital converter using successive comparisons|
|US5808575 *||Oct 31, 1996||Sep 15, 1998||Sony Corporation||Gain varying device capable of varying each gain of analog and digital signals|
|US5825318 *||Dec 19, 1996||Oct 20, 1998||Quantum Corporation||Data and servo sampling in synchronous data detection channel|
|US6172635 *||May 27, 1999||Jan 9, 2001||Toa Corporation||Highly accurate A/D converter|
|US6191725 *||Aug 30, 1999||Feb 20, 2001||Her Majesty The Queen In Right Of Canada, As Represented By Minister Of National Defence Of Her Majesty's Canadian Government||Automatic gain control for digital radar intercept receivers|
|US6333707 *||Dec 22, 1999||Dec 25, 2001||Nortel Networks Limited||Dynamic range extension of wideband receiver|
|EP0701745A1 *||Jun 6, 1994||Mar 20, 1996||Rca Thomson Licensing Corporation||Direct conversion tuner|
|1||Greger-Hansen et al., A Stacked A-To-D Converter for Increased Radar Signal Processor Dynamic Range, 2001 IEEE Radar Conference, May 2001, pp. 169-174.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US6801150 *||Feb 21, 2003||Oct 5, 2004||Denso Corporation||A/D conversion method and apparatus therefor|
|US6867717 *||Apr 4, 2003||Mar 15, 2005||Dalsa, Inc.||Digital encoder and method of encoding high dynamic range video images|
|US7965208||Aug 10, 2007||Jun 21, 2011||Entegris, Inc.||Auto ranging system and method for analog signal|
|US7984212 *||Jul 19, 2011||Lsi Corporation||System and method for utilizing first-in-first-out (FIFO) resources for handling differences in data rates between peripherals via a merge module that merges FIFO channels|
|US8490491||Feb 25, 2010||Jul 23, 2013||Olympus Ndt||Extended overflow indication for high dynamic range ultrasonic fault detection systems|
|US20030174082 *||Feb 21, 2003||Sep 18, 2003||Denso Corporation||A/D conversion method and apparatus therefor|
|US20070084288 *||Jul 20, 2006||Apr 19, 2007||Olympus Ndt||Ultrasonic fault detection system using a high dynamic range analog to digital conversion system|
|US20090178485 *||Jul 16, 2009||Olympus Ndt||Ultrasonic fault detection system using a high dynamic range analog to digital conversion system|
|US20090315745 *||Aug 10, 2007||Dec 24, 2009||Mcloughlin Robert F||Auto Ranging System and Method for Analog Signal|
|US20100268854 *||Apr 17, 2009||Oct 21, 2010||Sakthivel Komarasamy Pullagoundapatti||System and method for utilizing peripheral first-in-first-out (fifo) resources|
|U.S. Classification||341/155, 348/255, 341/139|
|Dec 17, 2001||AS||Assignment|
Owner name: HONEYWELL INTERNATIONAL INC., NEW JERSEY
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:NOLL, JOHN M.;BUNCH, BRIAN P.;MURTHY, SUNDARA;REEL/FRAME:012405/0612;SIGNING DATES FROM 20011214 TO 20011217
|Aug 6, 2007||REMI||Maintenance fee reminder mailed|
|Jan 27, 2008||LAPS||Lapse for failure to pay maintenance fees|
|Mar 18, 2008||FP||Expired due to failure to pay maintenance fee|
Effective date: 20080127